Concept dictionary
RLHF
RLHF stands for reinforcement learning from human feedback, a method that uses human preferences to help train a model toward more desirable responses.
- Review state
- Reviewed in POC
- Freshness
- Current
- Version
- 2
Also called
- RLHF
- Reinforcement Learning from Human Feedback
Remember it this wayHuman preferences train a reward; the reward steers the model.
In plain language
Reinforcement Learning from Human Feedback, or RLHF, uses human judgments to train a reward signal and then adjusts a model to produce outputs that score better under that signal.
Why it matters
RLHF can make outputs more useful or aligned with stated preferences, but the result reflects the reviewers, prompts, reward design, and optimization choices used in the process.
Example
Reviewers rank several answers to the same prompt. A reward model learns from the rankings, and reinforcement learning adjusts the language model toward higher-scoring responses.
Common misconception
RLHF does not give a model human values or eliminate harmful behavior. It optimizes against a learned approximation of selected human feedback.
Related terms
Sources
- Long Ouyang and colleagues, “Training Language Models to Follow Instructions with Human Feedback”documents supervised fine-tuning, reward-model training, and reinforcement-learning stages used in InstructGPT.
Learn more
- Compare [[alignment|Alignment]], [[human-review|Human Review]], [[fine-tuning|Fine-Tuning]], and [[direct-preference-optimization|Direct Preference Optimization]] to place RLHF in the surrounding system.
- Follow the cited source material above for the method, standard, or research context behind this draft.